Browse the corpus

INTRODUCTION Twin births have become more prevalent in developed countries over the last decades1, 2, 3. In 2014, the twin birth rate in the USA was 33.9 per 1000 live births, the highest rate ever recorded4. Twin gestations are at increased risk of maternal, perinatal and infant morbidity and mortality, as well as long‐term neurodevelopmental disability5, 6, 7, 8, 9, 10, 11, 12, 13. Moreover, twin gestations also have a significant impact on healthcare costs and quality of life for both the parents and the children7, 14, 15. Preterm birth is the most important factor determining neonatal morbidity and mortality among twins. The risk of preterm birth < 37 and < 32 weeks' gestation is eight‐ to ninefold higher in twin than in singleton gestations4. Several interventions have been proposed to reduce the rate of preterm birth in twin gestations, such as bed rest16, prophylactic tocolysis17, nutritional advice18, administration of 17α‐hydroxyprogesterone caproate19, vaginal progesterone19, cerclage20 and cervical pessary21, 22. Unfortunately, these interventions have not been shown to reduce the risk of preterm birth in unselected twin gestations.

n gestations, such as bed rest16, prophylactic tocolysis17, nutritional advice18, administration of 17α‐hydroxyprogesterone caproate19, vaginal progesterone19, cerclage20 and cervical pessary21, 22. Unfortunately, these interventions have not been shown to reduce the risk of preterm birth in unselected twin gestations. A short cervix, traditionally defined as a transvaginal sonographic cervical length (CL) ≤ 25 mm in the mid‐trimester of pregnancy, is an important risk factor for spontaneous preterm birth and has emerged as one of the strongest and most consistent predictors of preterm birth in asymptomatic women with singleton23, 24, 25, 26, 27, 28, 29 or twin gestations30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43. Currently, there is compelling evidence that administration of vaginal progesterone to asymptomatic women with a singleton gestation and a sonographic short cervix decreases the risk of preterm birth and neonatal morbidity and mortality44, 45, 46. The efficacy of vaginal progesterone in women with a twin gestation and a short cervix has been less studied.

ce that administration of vaginal progesterone to asymptomatic women with a singleton gestation and a sonographic short cervix decreases the risk of preterm birth and neonatal morbidity and mortality44, 45, 46. The efficacy of vaginal progesterone in women with a twin gestation and a short cervix has been less studied. A meta‐analysis of individual patient data (IPD) published in 2012 reported on the efficacy of vaginal progesterone in preventing preterm birth and neonatal morbidity and mortality in asymptomatic women with a twin gestation and a CL ≤ 25 mm in the mid‐trimester47. A total of 52 women (104 fetuses/infants) from three randomized controlled trials (RCTs) were included in the study. The use of vaginal progesterone was associated with a significant 44% reduction in the risk of composite neonatal morbidity and mortality (relative risk (RR), 0.56 (95% CI, 0.30–0.97)) and a 30% non‐significant reduction in the risk of preterm birth < 33 weeks' gestation (RR, 0.70 (95% CI, 0.34–1.44)). Since that time, additional RCTs evaluating the use of vaginal progesterone in twin gestations have been published. Therefore, a reassessment of the efficacy of this intervention in women with a twin gestation and a short cervix is justified. The objective of this study was to update the previous IPD meta‐analysis on the efficacy of vaginal progesterone in asymptomatic women with a twin gestation and a sonographic CL ≤ 25 mm in the mid‐trimester for the prevention of preterm birth and neonatal morbidity and mortality.

A meta‐analysis of individual patient data (IPD) published in 2012 reported on the efficacy of vaginal progesterone in preventing preterm birth and neonatal morbidity and mortality in asymptomatic women with a twin gestation and a CL ≤ 25 mm in the mid‐trimester47. A total of 52 women (104 fetuses/infants) from three randomized controlled trials (RCTs) were included in the study. The use of vaginal progesterone was associated with a significant 44% reduction in the risk of composite neonatal morbidity and mortality (relative risk (RR), 0.56 (95% CI, 0.30–0.97)) and a 30% non‐significant reduction in the risk of preterm birth < 33 weeks' gestation (RR, 0.70 (95% CI, 0.34–1.44)). Since that time, additional RCTs evaluating the use of vaginal progesterone in twin gestations have been published. Therefore, a reassessment of the efficacy of this intervention in women with a twin gestation and a short cervix is justified. The objective of this study was to update the previous IPD meta‐analysis on the efficacy of vaginal progesterone in asymptomatic women with a twin gestation and a sonographic CL ≤ 25 mm in the mid‐trimester for the prevention of preterm birth and neonatal morbidity and mortality. METHODS The study was conducted according to a prospectively prepared protocol and reported in accordance with the Preferred Reporting Items for Systematic reviews and Meta‐Analyses statement48. The review was registered with PROSPERO (number CRD42016039682).

The objective of this study was to update the previous IPD meta‐analysis on the efficacy of vaginal progesterone in asymptomatic women with a twin gestation and a sonographic CL ≤ 25 mm in the mid‐trimester for the prevention of preterm birth and neonatal morbidity and mortality. METHODS The study was conducted according to a prospectively prepared protocol and reported in accordance with the Preferred Reporting Items for Systematic reviews and Meta‐Analyses statement48. The review was registered with PROSPERO (number CRD42016039682). Data sources and searches We searched MEDLINE, EMBASE, POPLINE, CINAHL and LILACS (all from inception to 31 December 2016), the Cochrane Central Register of Controlled Trials and Research Registers of ongoing trials using a combination of keywords and text words related to ‘progesterone’, ‘preterm birth’, ‘randomized controlled trial’ and ‘twin gestation’. Google Scholar, proceedings of congresses on obstetrics, maternal‐fetal medicine and ultrasound in obstetrics, reference lists of identified studies, previously published systematic reviews and review articles were also searched. Experts in the field were contacted to identify further studies. No language restrictions were applied.

r, proceedings of congresses on obstetrics, maternal‐fetal medicine and ultrasound in obstetrics, reference lists of identified studies, previously published systematic reviews and review articles were also searched. Experts in the field were contacted to identify further studies. No language restrictions were applied. Study selection RCTs in which asymptomatic women with a twin gestation and a sonographic short cervix (CL ≤ 25 mm) in the mid‐trimester were allocated randomly to receive vaginal progesterone or placebo/no treatment for the prevention of preterm birth and/or adverse perinatal outcomes were eligible for inclusion in the review. Trials were included if the primary aim of the study was to prevent preterm birth in women with a twin gestation and a short cervix, or to prevent preterm birth in women with an unselected twin gestation but for whom outcomes were available in those with a prerandomization CL ≤ 25 mm. We excluded quasirandomized trials, trials that evaluated vaginal progesterone in women with preterm labor, arrested preterm labor (as maintenance tocolysis), preterm rupture of membranes or second‐trimester bleeding, trials that assessed vaginal progesterone in the first trimester only to prevent miscarriage and studies that did not report clinical outcomes. Studies published only as abstracts were excluded if additional information on methodological issues and results could not be obtained.

membranes or second‐trimester bleeding, trials that assessed vaginal progesterone in the first trimester only to prevent miscarriage and studies that did not report clinical outcomes. Studies published only as abstracts were excluded if additional information on methodological issues and results could not be obtained. All of the potentially relevant studies were retrieved and reviewed independently by two authors to determine inclusion. Disagreements were resolved by discussion amongst the reviewers. Data collection The corresponding author of each eligible trial was contacted and asked to provide anonymized data (without identifiers) about baseline characteristics and outcomes for every randomly assigned patient, as well as data on study characteristics and details of interventions and co‐interventions. All initial communications with authors were based on a template explaining the study and the data required. Data provided by the investigators were merged into a master database specifically constructed for the review. Data were checked for missing information, errors and inconsistencies by cross‐referencing with the publications of the original trials. Quality and integrity of the randomization processes were assessed by reviewing the chronological randomization sequence and pattern of assignment, as well as the balance of baseline characteristics across treatment groups. Inconsistencies or missing data were discussed with the authors and corrections were made when deemed necessary.

grity of the randomization processes were assessed by reviewing the chronological randomization sequence and pattern of assignment, as well as the balance of baseline characteristics across treatment groups. Inconsistencies or missing data were discussed with the authors and corrections were made when deemed necessary. Informed consent was provided by the patients upon enrollment in each of the original trials. In the present study, the data were not used for any purposes other than those of the original trial and no new data were collected. Therefore, informed consent specifically for this project was not considered necessary. This study was exempted from review by the Human Investigation Committee Administration Office of Wayne State University.

the data were not used for any purposes other than those of the original trial and no new data were collected. Therefore, informed consent specifically for this project was not considered necessary. This study was exempted from review by the Human Investigation Committee Administration Office of Wayne State University. Outcome measures The primary outcome measure was preterm birth < 33 weeks' gestation. Secondary outcome measures included: preterm birth < 37, < 36, < 35, < 34, < 32, < 30 and < 28 weeks' gestation; spontaneous preterm birth < 33 and < 34 weeks' gestation; respiratory distress syndrome (RDS); necrotizing enterocolitis; intraventricular hemorrhage; proven neonatal sepsis; retinopathy of prematurity; fetal death; neonatal death; perinatal death; a composite outcome of neonatal morbidity and mortality (defined as the occurrence of any of the following events: RDS, intraventricular hemorrhage, necrotizing enterocolitis, proven neonatal sepsis or neonatal death); birth weight < 1500 g and < 2500 g; admission to the neonatal intensive care unit; use of mechanical ventilation; and long‐term neurodevelopmental outcomes (suspected or diagnosed developmental delay, cerebral palsy, intellectual disabilities, vision impairment, hearing loss, cognitive and behavioral impairments and motor, communication and learning disorders at any age in childhood).

sive care unit; use of mechanical ventilation; and long‐term neurodevelopmental outcomes (suspected or diagnosed developmental delay, cerebral palsy, intellectual disabilities, vision impairment, hearing loss, cognitive and behavioral impairments and motor, communication and learning disorders at any age in childhood). Assessment of risk of bias The risk of bias in each included trial was assessed independently by two authors using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions 49. This tool assesses seven domains related to risk of bias (random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias) and categorizes studies by low, unclear or high risk of bias in each domain. Disagreements in risk of bias assessment were resolved through consensus.

generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias) and categorizes studies by low, unclear or high risk of bias in each domain. Disagreements in risk of bias assessment were resolved through consensus. Statistical analysis We included all randomized women and their fetuses/infants and performed all analyses on an intention‐to‐treat basis. For outcomes dealing with gestational age at delivery, the unit of analysis was the pregnancy, whereas for perinatal outcomes, the unit of analysis was the fetus/neonate. IPD were combined in a two‐stage approach in which outcomes were analyzed in the original trial and then summary statistics were generated using standard summary data meta‐analysis techniques to give an overall measure of effect (pooled RR with 95% CI)50. Heterogeneity of the results among studies was tested51 with the quantity I 2. We pooled results from individual studies using a fixed‐effect model if substantial statistical heterogeneity was not present (< 50%). If I 2 values were ≥ 50%, a random‐effects model was used to pool data across studies. For adverse perinatal outcomes, we estimated pooled RRs using analytical methods that assumed independence between neonates. However, to avoid incorrect conclusions due to the non‐independence of newborns from twin gestations, we also used a generalized linear model with generalized estimating equations to estimate parameters while controlling for cluster correlations52, 53, 54. The number needed to treat for benefit or harm, with a 95% CI, was calculated for outcomes for which there was a statistically significant reduction or increase in risk difference based on control event rates in the trials55.

imating equations to estimate parameters while controlling for cluster correlations52, 53, 54. The number needed to treat for benefit or harm, with a 95% CI, was calculated for outcomes for which there was a statistically significant reduction or increase in risk difference based on control event rates in the trials55. Subgroup analyses were performed to evaluate the effect of vaginal progesterone according to CL (<10, 10–20 and 21–25 mm), daily dose of vaginal progesterone (100, 200 and 400 mg) and obstetric history (no previous spontaneous preterm birth < 37 weeks' gestation and at least one previous spontaneous preterm birth < 37 weeks' gestation). A test for interaction between the treatment and subgroups was performed to examine whether treatment effects differed among subgroups56, 57, 58. An interaction P‐value ≥ 0.05 was considered to indicate that the effect of treatment did not differ significantly among subgroups. We planned to carry out sensitivity analyses to explore the effect of trial quality assessed by allocation concealment and random sequence generation (considering selection bias) and blinding (considering performance and detection biases), with studies rated as ‘high risk of bias’ or ‘unclear risk of bias’ for these domains being excluded from the analyses in order to assess whether this made any difference to the overall result. Subgroup and sensitivity analyses were only performed for the primary outcome of preterm birth < 33 weeks' gestation and for the secondary outcome of neonatal death. We also planned to explore potential sources of heterogeneity and to assess publication and related biases if at least 10 studies were included in a meta‐analysis, but these analyses were not undertaken due to the limited number of trials included in the review.

estation and for the secondary outcome of neonatal death. We also planned to explore potential sources of heterogeneity and to assess publication and related biases if at least 10 studies were included in a meta‐analysis, but these analyses were not undertaken due to the limited number of trials included in the review. Quality of evidence We used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, as outlined in the GRADE Handbook 59, to assess the quality of evidence for primary and secondary outcome measures. We considered evidence from RCTs as high quality but downgraded the evidence by one level for serious (or two levels for very serious) limitations based upon the following: design (risk of bias), consistency across studies, directness of the evidence, precision of estimates and presence of publication bias. The GRADEpro Guideline Development Tool60 was used to import data from Review Manager in order to create a ‘Summary of findings’ table to report the quality of the evidence. The GRADE approach results in an assessment of the quality of a body of evidence in one of four grades: (i) high: we are very confident that the true effect lies close to that of the estimate of the effect; (ii) moderate: we are moderately confident in the effect estimate, the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different; (iii) low: our confidence in the effect estimate is limited, the true effect may be substantially different from the estimate of the effect; and (iv) very low: we have very little confidence in the effect estimate, the true effect is likely to be substantially different from the estimate of effect.

tantially different; (iii) low: our confidence in the effect estimate is limited, the true effect may be substantially different from the estimate of the effect; and (iv) very low: we have very little confidence in the effect estimate, the true effect is likely to be substantially different from the estimate of effect. We performed all statistical analyses using Review Manager (RevMan, version 5.3.5; The Nordic Cochrane Centre, Copenhagen, Denmark) and SAS version 9.2 (SAS Institute, Cary, NC, USA) software. RESULTS Selection, characteristics and risk of bias of studies Figure 1 summarizes the process of identification and selection of studies. A total of 213 records were identified by the searches, of which nine were retrieved for full‐text review. Three studies, which evaluated vaginal progesterone in unselected twin gestations61, 62 or pregnancies conceived by in‐vitro fertilization or intracytoplasmic sperm injection63, were excluded because CL was not measured or collected before randomization or there were no data for women with a CL ≤ 25 mm at randomization. Six studies, including a total of 303 women (606 fetuses/infants) with a CL ≤ 25 mm, met the inclusion criteria64, 65, 66, 67, 68, 69; 159 women were assigned to vaginal progesterone and 144 to placebo/no treatment. Minimal differences were noted in baseline maternal characteristics between the vaginal progesterone and placebo/no treatment groups (Table 1). Figure 1 Study selection process. CL, cervical length. UOG-17397-FIG-0001-bTable 1 Baseline characteristics of pooled women

RESULTS Selection, characteristics and risk of bias of studies Figure 1 summarizes the process of identification and selection of studies. A total of 213 records were identified by the searches, of which nine were retrieved for full‐text review. Three studies, which evaluated vaginal progesterone in unselected twin gestations61, 62 or pregnancies conceived by in‐vitro fertilization or intracytoplasmic sperm injection63, were excluded because CL was not measured or collected before randomization or there were no data for women with a CL ≤ 25 mm at randomization. Six studies, including a total of 303 women (606 fetuses/infants) with a CL ≤ 25 mm, met the inclusion criteria64, 65, 66, 67, 68, 69; 159 women were assigned to vaginal progesterone and 144 to placebo/no treatment. Minimal differences were noted in baseline maternal characteristics between the vaginal progesterone and placebo/no treatment groups (Table 1). Figure 1 Study selection process. CL, cervical length. UOG-17397-FIG-0001-bTable 1 Baseline characteristics of pooled women Characteristic Vaginal progesterone (n = 159) Placebo/no treatment (n = 144) Maternal age (years) 27 (25–30) 28 (25–31) Body mass index (kg/m2) 22.4 (21.2–25.7)* 22.9 (21.0–25.4)† Smoker 3 (1.9) 3 (2.1) Previous spontaneous PTB 28 (17.6) 28 (19.4) Monochorionic pregnancy 8 (5.0) 6 (4.2) GA at randomization (weeks) 21.7 (20.6–23.1) 22.1 (21.1–23.3) CL at randomization (mm) 22 (20–23) 22 (20–23) CL ≤ 20 mm at randomization 49 (30.8) 47 (32.6) Data are given as median (interquartile range) or n (%). * n = 41. † n = 36.

Smoker 3 (1.9) 3 (2.1) Previous spontaneous PTB 28 (17.6) 28 (19.4) Monochorionic pregnancy 8 (5.0) 6 (4.2) GA at randomization (weeks) 21.7 (20.6–23.1) 22.1 (21.1–23.3) CL at randomization (mm) 22 (20–23) 22 (20–23) CL ≤ 20 mm at randomization 49 (30.8) 47 (32.6) Data are given as median (interquartile range) or n (%). * n = 41. † n = 36. CL, cervical length; GA, gestational age; PTB, preterm birth. The individual characteristics of the studies included in the meta‐analysis are shown in Table 2. Five studies were double‐blind, placebo‐controlled trials64, 65, 66, 67, 68. The remaining study compared vaginal progesterone with no treatment69. Three studies were performed in low/middle‐income countries65, 68, 69, two in high‐income countries66, 67 and one in both low/middle‐ and high‐income countries64. Two trials were specifically designed to evaluate the use of vaginal progesterone in women with a twin gestation and a sonographic short cervix (CL ≤ 15 mm64 and CL between 20 and 25 mm69). The remaining four studies tested the effect of vaginal progesterone in women with unselected twin gestations and their authors provided data relevant to women with a CL ≤ 25 mm before randomization65, 66, 67, 68. The trial that assessed vaginal progesterone in women with a CL between 20 and 25 mm69 provided data for 224 mothers and their 448 fetuses/infants. The other five studies provided data for 79 women and 158 fetuses/infants. Table 2 Characteristics of studies included in the systematic review

The individual characteristics of the studies included in the meta‐analysis are shown in Table 2. Five studies were double‐blind, placebo‐controlled trials64, 65, 66, 67, 68. The remaining study compared vaginal progesterone with no treatment69. Three studies were performed in low/middle‐income countries65, 68, 69, two in high‐income countries66, 67 and one in both low/middle‐ and high‐income countries64. Two trials were specifically designed to evaluate the use of vaginal progesterone in women with a twin gestation and a sonographic short cervix (CL ≤ 15 mm64 and CL between 20 and 25 mm69). The remaining four studies tested the effect of vaginal progesterone in women with unselected twin gestations and their authors provided data relevant to women with a CL ≤ 25 mm before randomization65, 66, 67, 68. The trial that assessed vaginal progesterone in women with a CL between 20 and 25 mm69 provided data for 224 mothers and their 448 fetuses/infants. The other five studies provided data for 79 women and 158 fetuses/infants. Table 2 Characteristics of studies included in the systematic review Study Country Primary target population Inclusion and exclusion criteria Women with CL ≤ 25 mm (n)/fetuses or infants (n) Intervention Primary outcome measure Fonseca (2007)64 UK, Chile, Brazil, Greece Women with short cervix Inclusion: women with singleton or twin gestation and transvaginal sonographic CL ≤ 15 mm

Table 2 Characteristics of studies included in the systematic review Study Country Primary target population Inclusion and exclusion criteria Women with CL ≤ 25 mm (n)/fetuses or infants (n) Intervention Primary outcome measure Fonseca (2007)64 UK, Chile, Brazil, Greece Women with short cervix Inclusion: women with singleton or twin gestation and transvaginal sonographic CL ≤ 15 mm Exclusion: major fetal abnormality, painful regular uterine contractions, history of ruptured membranes or cervical cerclage 24/48 Vaginal progesterone capsule (200 mg/day) or placebo from 24 to 33 + 6 weeks Spontaneous PTB < 34 weeks Cetingoz (2011)65 Turkey Women at high risk of PTB Inclusion: women with at least one previous spontaneous PTB, uterine malformation or twin gestation Exclusion: in‐place or planned cervical cerclage or serious fetal anomaly 7/14 Vaginal progesterone suppository (100 mg/day) or placebo from 24 to 34 weeks PTB < 37 weeks Rode (2011)66 Denmark, Austria Women with twin gestation Inclusion: women with a diamniotic twin gestation and chorionicity assessed by ultrasound before 16 weeks

‐place or planned cervical cerclage or serious fetal anomaly 7/14 Vaginal progesterone suppository (100 mg/day) or placebo from 24 to 34 weeks PTB < 37 weeks Rode (2011)66 Denmark, Austria Women with twin gestation Inclusion: women with a diamniotic twin gestation and chorionicity assessed by ultrasound before 16 weeks Exclusion: higher order multiple pregnancies, age < 18 years, known allergy to progesterone or peanuts as active treatment contained peanut oil, history of hormone‐associated thromboembolic disorders, rupture of membranes, pregnancies treated for or with signs of TTTS, intentional fetal reduction, known major structural or chromosomal fetal abnormality, known or suspected malignancy in genitals or breasts or known liver disease 21/42 Vaginal progesterone pessary (200 mg/day) or placebo from 20 to 23 + 6 up to 33 + 6 weeks PTB < 34 weeks Serra (2013)67 Spain Women with twin gestation Inclusion: women with dichorionic diamniotic twin gestation

mosomal fetal abnormality, known or suspected malignancy in genitals or breasts or known liver disease 21/42 Vaginal progesterone pessary (200 mg/day) or placebo from 20 to 23 + 6 up to 33 + 6 weeks PTB < 34 weeks Serra (2013)67 Spain Women with twin gestation Inclusion: women with dichorionic diamniotic twin gestation Exclusion: monochorionic twin gestation, triplet or higher order multiple gestation, elective cervical cerclage prior to 14 weeks, history of hepatic problem or gestational cholestasis, abnormal liver enzymes, abnormal kidney function, local allergy to micronized natural progesterone or peanuts, recurrent vaginal bleeding, recurrent vaginal infection, fetal anomaly, alcohol or illicit drug consumption or smoking ≥ 10 cigarettes/day 6/12 Vaginal progesterone pessary (200 or 400 mg/day) or placebo from 20 to 34 weeks PTB < 37 weeks Brizot (2015)68 Brazil Women with twin gestation Inclusion: women with naturally conceived diamniotic twin gestation, no previous PTB and gestational age between 18 + 0 and 21 + 6 weeks

ion or smoking ≥ 10 cigarettes/day 6/12 Vaginal progesterone pessary (200 or 400 mg/day) or placebo from 20 to 34 weeks PTB < 37 weeks Brizot (2015)68 Brazil Women with twin gestation Inclusion: women with naturally conceived diamniotic twin gestation, no previous PTB and gestational age between 18 + 0 and 21 + 6 weeks Exclusion: major fetal abnormality, allergy to progesterone or peanuts, hepatic dysfunction, porphyria, otosclerosis, malignant disease, severe depressive state, current or previous thromboembolic disease, uterine malformation, prophylactic cerclage or ovular infection 21/42 Vaginal progesterone ovule (200 mg/day) or placebo from 18 to 21 + 6 up to 34 + 6 weeks Mean gestational age at delivery El‐Refaie (2016)69 Egypt Women with twin gestation and short cervix Inclusion: women with dichorionic twin gestation, gestational age between 20 and 24 weeks, transvaginal sonographic CL between 20 and 25 mm, and without signs or symptoms of preterm labor Exclusion: known allergy or contraindication to progesterone therapy, monochorionic twin gestation, known major fetal structural or chromosomal abnormality, single fetal demise, fetal reduction in current pregnancy, cervical cerclage in current pregnancy, medical conditions that may lead to preterm labor, rupture of membranes or vaginal bleeding 224/448 Vaginal progesterone suppository (400 mg/day) from 20 to 24 up to 37 weeks or no treatment PTB < 34 weeks Only the first author of each study is given. CL, cervical length; PTB, preterm birth; TTTS, twin‐to‐twin transfusion syndrome.

Exclusion: known allergy or contraindication to progesterone therapy, monochorionic twin gestation, known major fetal structural or chromosomal abnormality, single fetal demise, fetal reduction in current pregnancy, cervical cerclage in current pregnancy, medical conditions that may lead to preterm labor, rupture of membranes or vaginal bleeding 224/448 Vaginal progesterone suppository (400 mg/day) from 20 to 24 up to 37 weeks or no treatment PTB < 34 weeks Only the first author of each study is given. CL, cervical length; PTB, preterm birth; TTTS, twin‐to‐twin transfusion syndrome. Three studies used vaginal progesterone 200 mg/day (capsule64, pessary66 or ovule68), one used vaginal progesterone suppositories 100 mg/day65, one used vaginal progesterone suppositories 400 mg/day69 and the remaining study used vaginal progesterone suppositories 200 or 400 mg/day67. Treatment was started between 20 and 24 weeks' gestation in five trials64, 65, 66, 67, 69, and between 18 and 21 weeks' gestation in the remaining trial68. Five studies reported that participants received medication from the time of enrollment until ∼34 weeks' gestation64, 65, 66, 67, 68, and one study reported medication from enrollment until 37 weeks' gestation69. Two trials included only women with a dichorionic twin gestation67, 69. Major fetal abnormality, cervical cerclage in place or planned, allergy to progesterone and hepatic dysfunction were reported as exclusion criteria in most studies. The primary outcome measure was preterm birth < 34 weeks' gestation in two trials66, 69, preterm birth < 37 weeks' gestation in two trials65, 67, spontaneous preterm birth < 34 weeks' gestation in one trial64 and mean gestational age at delivery in the remaining study68. The study by El‐Refaie et al. 69 did not collect data for some neonatal morbidities, such as necrotizing enterocolitis, intraventricular hemorrhage, proven neonatal sepsis and retinopathy of prematurity.

us preterm birth < 34 weeks' gestation in one trial64 and mean gestational age at delivery in the remaining study68. The study by El‐Refaie et al. 69 did not collect data for some neonatal morbidities, such as necrotizing enterocolitis, intraventricular hemorrhage, proven neonatal sepsis and retinopathy of prematurity. The risk of bias in each included study is summarized in Figure 2. All studies had adequate generation of allocation sequence and concealment of allocation, and appeared to be free of selective outcome reporting and other sources of bias. Five studies were considered to be at low risk of selection, performance, detection, attrition and reporting biases64, 65, 66, 67, 68. The study by El‐Refaie et al. 69 had a high risk of performance and detection biases because patients, clinical staff and outcome assessors were not blinded to the allocated interventions. In addition, this trial was judged to be at unclear risk of attrition bias because the number of losses to follow‐up was not balanced across study groups (7.2% in the vaginal progesterone group and 13.6% in the no treatment group). Figure 2 Risk of bias of studies included in the systematic review. , low risk of bias; , high risk of bias; , unclear risk of bias.

The risk of bias in each included study is summarized in Figure 2. All studies had adequate generation of allocation sequence and concealment of allocation, and appeared to be free of selective outcome reporting and other sources of bias. Five studies were considered to be at low risk of selection, performance, detection, attrition and reporting biases64, 65, 66, 67, 68. The study by El‐Refaie et al. 69 had a high risk of performance and detection biases because patients, clinical staff and outcome assessors were not blinded to the allocated interventions. In addition, this trial was judged to be at unclear risk of attrition bias because the number of losses to follow‐up was not balanced across study groups (7.2% in the vaginal progesterone group and 13.6% in the no treatment group). Figure 2 Risk of bias of studies included in the systematic review. , low risk of bias; , high risk of bias; , unclear risk of bias. UOG-17397-FIG-0002-cEffect of vaginal progesterone on preterm birth Women allocated to receive vaginal progesterone had a significantly lower risk of preterm birth < 33 weeks' gestation (31.4% vs 43.1%; RR, 0.69 (95% CI, 0.51– 0.93); P = 0.01; I 2 = 0%; six studies, 303 women; moderate‐quality evidence) compared with those allocated to placebo/no treatment (Figure 3). In addition, vaginal progesterone was associated with a significant reduction in the risk of preterm birth < 35 weeks' gestation (RR, 0.83 (95% CI, 0.69–0.99); moderate‐quality evidence), < 34 weeks' gestation (RR, 0.71 (95% CI, 0.56–0.91); moderate‐quality evidence), < 32 weeks' gestation (RR, 0.51 (95% CI, 0.34–0.77); moderate‐quality evidence), < 30 weeks' gestation (RR, 0.47 (95% CI, 0.25–0.86); moderate‐quality evidence), and spontaneous preterm birth at < 33 weeks' gestation (RR, 0.67 (95% CI, 0.48–0.93); moderate‐quality evidence) and < 34 weeks' gestation (RR, 0.71 (95% CI, 0.54–0.93); moderate‐quality evidence) (Table 3). The number needed to treat to prevent one case of preterm birth occurring at < 30 to < 35 gestational weeks varied from 6 to 12. There were no significant differences between the study groups in the risk of preterm birth < 37 weeks' (moderate‐quality evidence), < 36 weeks' (moderate‐quality evidence) and < 28 weeks' (low‐quality evidence) gestation.

revent one case of preterm birth occurring at < 30 to < 35 gestational weeks varied from 6 to 12. There were no significant differences between the study groups in the risk of preterm birth < 37 weeks' (moderate‐quality evidence), < 36 weeks' (moderate‐quality evidence) and < 28 weeks' (low‐quality evidence) gestation. Figure 3 Forest plot of the effect of vaginal progesterone on the risk of preterm birth < 33 weeks' gestation. CI, confidence interval. UOG-17397-FIG-0003-cTable 3 Effect of vaginal progesterone on the risk of preterm birth Events (n)/Total (N) Outcome Trials (n refs) Vaginal progesterone Placebo/no treatment Pooled RR (95% CI) I 2 (%) NNT (95% CI) Preterm birth < 37 weeks 664, 65, 66, 67, 68, 69 137/159 131/144 0.94 (0.86–1.02) 0 — Preterm birth < 36 weeks 664, 65, 66, 67, 68, 69 112/159 110/144 0.92 (0.80–1.05) 0 — Preterm birth < 35 weeks 664, 65, 66, 67, 68, 69 90/159 98/144 0.83 (0.69–0.99) 0 9 (5–147) Preterm birth < 34 weeks 664, 65, 66, 67, 68, 69 63/159 78/144 0.71 (0.56–0.91) 0 6 (4–21) Preterm birth < 32 weeks 664, 65, 66, 67, 68, 69 29/159 46/144 0.51 (0.34–0.77) 0 6 (5–14) Preterm birth < 30 weeks 664, 65, 66, 67, 68, 69 14/159 22/144 0.47 (0.25–0.86) 0 12 (9–47) Preterm birth < 28 weeks 664, 65, 66, 67, 68, 69 9/159 12/144 0.51 (0.24–1.08) 0 — Spontaneous preterm birth < 33 weeks 664, 65, 66, 67, 68, 69 42/159 54/144 0.67 (0.48–0.93) 0 8 (5–38) Spontaneous preterm birth < 34 weeks 664, 65, 66, 67, 68, 69 55/159 69/144 0.71 (0.54–0.93) 0 7 (5–30) CI, confidence interval; NNT, number needed to treat; refs, reference numbers; RR, relative risk.

(0.24–1.08) 0 — Spontaneous preterm birth < 33 weeks 664, 65, 66, 67, 68, 69 42/159 54/144 0.67 (0.48–0.93) 0 8 (5–38) Spontaneous preterm birth < 34 weeks 664, 65, 66, 67, 68, 69 55/159 69/144 0.71 (0.54–0.93) 0 7 (5–30) CI, confidence interval; NNT, number needed to treat; refs, reference numbers; RR, relative risk. Effect of vaginal progesterone on adverse perinatal outcomes Infants whose mothers received vaginal progesterone had a significantly lower risk of neonatal death (RR, 0.53 (95% CI, 0.35–0.81); moderate‐quality evidence), perinatal death (RR, 0.58 (95% CI, 0.39–0.84); moderate‐quality evidence), RDS (RR, 0.70 (95% CI, 0.56–0.89); moderate‐quality evidence), composite neonatal morbidity and mortality (RR, 0.61 (95% CI, 0.34–0.98); moderate‐quality evidence), birth weight < 1500 g (RR, 0.53 (95% CI, 0.35–0.80); moderate‐quality evidence) and use of mechanical ventilation (RR, 0.54 (95% CI, 0.36–0.81); moderate‐quality evidence) (Table 4). The number needed to treat to prevent one case of these adverse perinatal outcomes varied from 6 to 8. There was no evidence of an effect of vaginal progesterone on necrotizing enterocolitis (low‐quality evidence), intraventricular hemorrhage (low‐quality evidence), proven neonatal sepsis (low‐quality evidence), retinopathy of prematurity (low‐quality evidence), fetal death (very low‐quality evidence), birth weight < 2500 g (moderate‐quality evidence) and admission to the neonatal intensive care unit (moderate‐quality evidence). Table 4 Effect of vaginal progesterone on the risk of adverse perinatal outcomes

Effect of vaginal progesterone on adverse perinatal outcomes Infants whose mothers received vaginal progesterone had a significantly lower risk of neonatal death (RR, 0.53 (95% CI, 0.35–0.81); moderate‐quality evidence), perinatal death (RR, 0.58 (95% CI, 0.39–0.84); moderate‐quality evidence), RDS (RR, 0.70 (95% CI, 0.56–0.89); moderate‐quality evidence), composite neonatal morbidity and mortality (RR, 0.61 (95% CI, 0.34–0.98); moderate‐quality evidence), birth weight < 1500 g (RR, 0.53 (95% CI, 0.35–0.80); moderate‐quality evidence) and use of mechanical ventilation (RR, 0.54 (95% CI, 0.36–0.81); moderate‐quality evidence) (Table 4). The number needed to treat to prevent one case of these adverse perinatal outcomes varied from 6 to 8. There was no evidence of an effect of vaginal progesterone on necrotizing enterocolitis (low‐quality evidence), intraventricular hemorrhage (low‐quality evidence), proven neonatal sepsis (low‐quality evidence), retinopathy of prematurity (low‐quality evidence), fetal death (very low‐quality evidence), birth weight < 2500 g (moderate‐quality evidence) and admission to the neonatal intensive care unit (moderate‐quality evidence). Table 4 Effect of vaginal progesterone on the risk of adverse perinatal outcomes Pooled RR (95% CI) Events (n)/Total (N) Outcome Trials (n refs) Vaginal progesterone Placebo/no treatment Assuming independence between twins Adjustment for non‐independence between twins I 2 (%) NNT (95% CI) Respiratory distress syndrome 664, 65, 66, 67, 68, 69 102/311 131/280 0.67 (0.55–0.82) 0.70 (0.56–0.89) 0 6 (4–16) Necrotizing enterocolitis 564, 65, 66, 67, 68 1/82 0/68 1.00 (0.04–22.43) 1.07 (0.05–22.25) NA — Intraventricular hemorrhage 564, 65, 66, 67, 68 2/80 2/68 0.93 (0.15–5.75) 1.47 (0.22–9.63) 0 — Proven neonatal sepsis 564, 65, 66, 67, 68 4/80 7/68 0.44 (0.13–1.46) 0.59 (0.18–1.93) 0 — Retinopathy of prematurity 564, 65, 66, 67, 68 1/80 1/68 0.42 (0.07–2.56) 0.45 (0.08–2.59) 17 — Fetal death 664, 65, 66, 67, 68, 69 9/318 9/288 0.57 (0.23–1.42) 0.68 (0.26–1.84) 0 — Neonatal death 664, 65, 66, 67, 68, 69 34/318 63/288 0.50 (0.34–0.71) 0.53 (0.35–0.81) 25 8 (5–19) Perinatal death 664, 65, 66, 67, 68, 69 43/318 72/288 0.51 (0.36–0.70) 0.58 (0.39–0.84) 24 7 (5–20) Composite neonatal morbidity/mortality* 564, 65, 66, 67, 68 23/84 28/70 0.57 (0.36–0.93) 0.61 (0.34–0.98) 0 6 (3–109) Birth weight < 1500 g 664, 65, 66, 67, 68, 69 48/315 73/280 0.52 (0.38–0.72) 0.53 (0.35–0.80) 17 7 (5–17) Birth weight < 2500 g 664, 65, 66, 67, 68, 69 244/315 223/280 0.97 (0.89–1.06) 0.99 (0.89–1.10) 0 — Admission to the NICU 664, 65, 66, 67, 68, 69 211/315 209/282 0.92 (0.83–1.02) 0.95 (0.84–1.08) 0 — Mechanical ventilation 664, 65, 66, 67, 68, 69 49/311 76/280 0.52 (0.37–0.71) 0.54 (0.36–0.81) 0 7 (5–17) * Occurrence of any of the following events: respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, proven neonatal sepsis or neonatal death.

82 0.92 (0.83–1.02) 0.95 (0.84–1.08) 0 — Mechanical ventilation 664, 65, 66, 67, 68, 69 49/311 76/280 0.52 (0.37–0.71) 0.54 (0.36–0.81) 0 7 (5–17) * Occurrence of any of the following events: respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, proven neonatal sepsis or neonatal death. CI, confidence interval; NA, not applicable; NICU, neonatal intensive care unit; NNT, number needed to treat; refs, reference numbers; RR, relative risk. Subgroup and sensitivity analyses Subgroup analyses of the effect of vaginal progesterone on preterm birth < 33 weeks' gestation and neonatal death, according to CL, daily dose of vaginal progesterone and obstetric history, are shown in Table 5. There was no evidence that women in any one of the prespecified subgroups benefited more or less from the use of vaginal progesterone than those in any other subgroup (all, interaction P‐value ≥ 0.40). Nonetheless, vaginal progesterone was associated with a statistically significant reduction in the risk of preterm birth < 33 weeks' gestation and neonatal death in women with a CL between 10 and 20 mm (RR, 0.44 (95% CI, 0.22–0.87) and 0.20 (95% CI, 0.05–0.86), respectively) and women who were administered 400 mg of daily vaginal progesterone (RR, 0.66 (95% CI, 0.46–0.95) and 0.42 (95% CI, 0.23–0.76), respectively). Moreover, vaginal progesterone significantly decreased the risk of neonatal death in women with a CL between 21 and 25 mm (RR, 0.57 (95% CI, 0.36–0.90)) and women with no previous spontaneous preterm birth (RR, 0.58 (95% CI, 0.36–0.93)).

sterone (RR, 0.66 (95% CI, 0.46–0.95) and 0.42 (95% CI, 0.23–0.76), respectively). Moreover, vaginal progesterone significantly decreased the risk of neonatal death in women with a CL between 21 and 25 mm (RR, 0.57 (95% CI, 0.36–0.90)) and women with no previous spontaneous preterm birth (RR, 0.58 (95% CI, 0.36–0.93)). Table 5 Subgroup analyses of the effect of vaginal progesterone on preterm birth < 33 weeks' gestation and neonatal death Preterm birth < 33 weeks' gestation Neonatal death Subgroup n Pooled RR (95% CI) Interaction P‐value n Pooled RR (95% CI)* Interaction P‐value Cervical length 0.40 0.40 < 10 mm 14 0.74 (0.37–1.49) 28 0.67 (0.12–3.70) 10–20 mm 82 0.44 (0.22–0.87) 164 0.20 (0.05–0.86) 21–25 mm 207 0.74 (0.51–1.06) 414 0.57 (0.36–0.90) Daily dose of vaginal progesterone 0.77 0.60 100 mg 7 0.40 (0.04–3.74) 14 0.09 (0.00–3.59) 200 mg 69 0.79 (0.48–1.30) 138 0.66 (0.15–2.86) 400 mg 227 0.66 (0.46–0.95) 454 0.42 (0.23–0.76) Obstetric history 0.40 0.62 No previous preterm birth 247 0.72 (0.52–1.01) 494 0.58 (0.36–0.93) ≥ 1 previous preterm birth 56 0.50 (0.22–1.11) 112 0.45 (0.18–1.10) * Adjusted for non‐independence between twins. CI, confidence interval; RR, relative risk.

Preterm birth < 33 weeks' gestation Neonatal death Subgroup n Pooled RR (95% CI) Interaction P‐value n Pooled RR (95% CI)* Interaction P‐value Cervical length 0.40 0.40 < 10 mm 14 0.74 (0.37–1.49) 28 0.67 (0.12–3.70) 10–20 mm 82 0.44 (0.22–0.87) 164 0.20 (0.05–0.86) 21–25 mm 207 0.74 (0.51–1.06) 414 0.57 (0.36–0.90) Daily dose of vaginal progesterone 0.77 0.60 100 mg 7 0.40 (0.04–3.74) 14 0.09 (0.00–3.59) 200 mg 69 0.79 (0.48–1.30) 138 0.66 (0.15–2.86) 400 mg 227 0.66 (0.46–0.95) 454 0.42 (0.23–0.76) Obstetric history 0.40 0.62 No previous preterm birth 247 0.72 (0.52–1.01) 494 0.58 (0.36–0.93) ≥ 1 previous preterm birth 56 0.50 (0.22–1.11) 112 0.45 (0.18–1.10) * Adjusted for non‐independence between twins. CI, confidence interval; RR, relative risk. When the sensitivity analysis was restricted to the five trials with adequate blinding of patients, clinical staff and outcome assessors64, 65, 66, 67, 68, the effect of vaginal progesterone on the reduction in the risk of preterm birth < 33 weeks' gestation and neonatal death was non‐significant (RR, 0.77 (95% CI, 0.48–1.24) and 0.56 (95% CI, 0.21–1.48), respectively). However, it should be noted that the sensitivity analyses did not substantially change the magnitude and direction of effect sizes obtained in the overall analyses. Sensitivity analyses based on allocation concealment and random sequence generation were not performed because there were no trials at unclear or high risk of bias for these domains.

that the sensitivity analyses did not substantially change the magnitude and direction of effect sizes obtained in the overall analyses. Sensitivity analyses based on allocation concealment and random sequence generation were not performed because there were no trials at unclear or high risk of bias for these domains. Effect of vaginal progesterone on long‐term neurodevelopmental outcomes No study has reported the effects of vaginal progesterone on long‐term neurodevelopmental outcomes in twin gestations with a short cervix. Thus far, two trials have reported the effects of prenatal exposure to vaginal progesterone on long‐term neurodevelopmental outcomes in unselected twin gestations70, 71. In 2015, a follow‐up study of one of the excluded trials61 reported that there was no significant difference in developmental delay (assessed using the Child Development Inventory tool) between twins exposed to either vaginal progesterone (42/140) or placebo (65/184) at a mean age of 55.5 months (odds ratio (OR), 0.87 (95% CI, 0.46–1.63))70. Recently, one of the studies included in the review66 reported on the developmental performance of children exposed prenatally to vaginal progesterone (n = 225) or placebo (n = 212), at a mean age of 57 months71. The developmental performance was evaluated by the parent‐completed Ages and Stages Questionnaire (ASQ) screening tool. Overall, mean ASQ total scores were significantly higher in the vaginal progesterone‐exposed group (269 ± 28) than in the placebo‐exposed group (262 ± 31) (P = 0.03), although there was no statistically significant difference in the risk of low ASQ score (< 10th percentile) between the study groups (OR, 0.47 (95% CI, 0.21–1.06)). A subgroup analysis showed that dichorionic twins who were exposed prenatally to vaginal progesterone had a significantly lower risk of having a low total ASQ score than those who were exposed to placebo (OR, 0.34 (95% CI, 0.14–0.86)).

Q score (< 10th percentile) between the study groups (OR, 0.47 (95% CI, 0.21–1.06)). A subgroup analysis showed that dichorionic twins who were exposed prenatally to vaginal progesterone had a significantly lower risk of having a low total ASQ score than those who were exposed to placebo (OR, 0.34 (95% CI, 0.14–0.86)). DISCUSSION Principal findings The main finding in this updated IPD meta‐analysis is that the administration of vaginal progesterone to asymptomatic women with a twin gestation and a mid‐trimester sonographic short cervix significantly reduces the risk of preterm birth < 33 weeks' gestation (primary outcome) by 31% and neonatal death by 47%. In addition, patients who received vaginal progesterone had a significantly decreased risk of preterm birth < 35, < 34, < 32 and < 30 weeks, spontaneous preterm birth < 33 and < 34 weeks, perinatal death, composite neonatal morbidity and mortality, RDS, birth weight < 1500 g and use of mechanical ventilation. Moreover, evidence from two trials that assessed vaginal progesterone in unselected twin gestations showed that there were no significant differences in the risk of neurodevelopmental disability at 4–5 years of age between children exposed prenatally to vaginal progesterone and those exposed to placebo.

al ventilation. Moreover, evidence from two trials that assessed vaginal progesterone in unselected twin gestations showed that there were no significant differences in the risk of neurodevelopmental disability at 4–5 years of age between children exposed prenatally to vaginal progesterone and those exposed to placebo. Quality of the evidence Evidence for most critical outcomes assessed with GRADE methodology was considered to be of moderate quality (Table S1). We downgraded the evidence from high quality to moderate quality because most of the pooled effect was provided by one study with moderate risk of bias. A judgment of moderate quality means that we have some confidence that our results approach the true impact of vaginal progesterone on preterm birth and adverse neonatal outcomes in twin gestations with a short cervix; at the same time, we acknowledge that future trials may change these results.

moderate risk of bias. A judgment of moderate quality means that we have some confidence that our results approach the true impact of vaginal progesterone on preterm birth and adverse neonatal outcomes in twin gestations with a short cervix; at the same time, we acknowledge that future trials may change these results. Subgroup analyses We evaluated several clinically important subgroups based on CL, daily dose of vaginal progesterone and history of spontaneous preterm birth. Overall, subgroup analyses indicated that the beneficial effects of vaginal progesterone did not differ significantly across patient groups, as the interaction tests for subgroup differences were non‐significant. Patients with a CL between 10 and 20 mm or those who received vaginal progesterone 400 mg/day seemed to have a greater‐than‐average reduction in the risk of preterm birth < 33 weeks' gestation and neonatal death. However, analyses of categories such as CL < 10 mm, daily dose of vaginal progesterone of 100 or 200 mg and history of spontaneous preterm birth were based on small numbers of women, reflecting the pattern of recruitment to the original trials, in which most women had a CL between 10 and 25 mm, used vaginal progesterone 400 mg/day and did not have a history of spontaneous preterm birth. As a result, our analysis was limited in its power to estimate effects within those groups of patients. Thus, although prespecified and clinically interesting, these subgroup analyses should be interpreted with caution.

25 mm, used vaginal progesterone 400 mg/day and did not have a history of spontaneous preterm birth. As a result, our analysis was limited in its power to estimate effects within those groups of patients. Thus, although prespecified and clinically interesting, these subgroup analyses should be interpreted with caution. Lack of long‐term adverse neurodevelopmental outcomes in twins exposed to vaginal progesterone during pregnancy Current evidence suggests that in‐utero exposure to vaginal progesterone, administered in twin gestations for the prevention of preterm birth, has no detrimental effects on long‐term neurodevelopmental outcomes. A total of 761 surviving children who participated in two placebo‐controlled trials of vaginal progesterone to prevent preterm birth in unselected twin gestations61, 66 were evaluated at a mean age of ∼56 months for neurodevelopmental outcomes70, 71. Both studies reported no significant differences in the risk of developmental delay70 or suspected developmental delay71 between children whose mothers received vaginal progesterone and those whose mothers received placebo. It should be noted that vaginal progesterone had no effect on gestational age at delivery in both trials, which allowed the assessment of the direct effect of vaginal progesterone on childhood neurodevelopmental outcomes independent of any effect of vaginal progesterone on preterm birth. Interestingly, a subgroup analysis of one of these studies70 found that dichorionic twins who were exposed prenatally to progesterone had a significantly reduced risk of a low total ASQ score, a higher total mean ASQ score and higher mean ASQ scores in communication, gross motor skills and personal/social skills in comparison with dichorionic twins who were exposed to placebo. These findings suggest a potential long‐term benefit related to prenatal exposure to vaginal progesterone, which would not be surprising because there is some evidence indicating that progesterone could act as a neuroprotectant for brain disorders, mainly traumatic brain injury72. Thereby, a direct beneficial effect of vaginal progesterone on childhood neurodevelopment would be plausible. This issue deserves further investigation.

ould not be surprising because there is some evidence indicating that progesterone could act as a neuroprotectant for brain disorders, mainly traumatic brain injury72. Thereby, a direct beneficial effect of vaginal progesterone on childhood neurodevelopment would be plausible. This issue deserves further investigation. Lack of long‐term adverse health outcomes in twins exposed to vaginal progesterone during pregnancy With regard to the effects of the prenatal exposure to vaginal progesterone on childhood health outcomes in twins, the follow‐up study by McNamara et al.70 reported that there were no significant differences between vaginal progesterone‐exposed and placebo‐exposed twins with respect to death, congenital malformations, growth, health service utilization and global health status at 3–6 years of age. The follow‐up study by Vedel et al.71 reported that the rates of diagnoses related to 10 organ systems, the median number of hospital admissions and the median length of hospital stay did not differ significantly between the vaginal progesterone‐ and placebo‐exposed twins up to 8 years of age. Notwithstanding, in subgroup analyses restricted to dichorionic twins and diagnoses made solely during hospital admission, the investigators found that diagnoses related to structural and functional abnormalities of the heart were significantly more frequent among children who were exposed prenatally to vaginal progesterone. However, these differences became non‐significant after Bonferroni adjustment for multiple comparisons. In conclusion, second‐ and third‐trimester exposure to vaginal progesterone does not seem to have harmful effects on the childhood health of twins.

among children who were exposed prenatally to vaginal progesterone. However, these differences became non‐significant after Bonferroni adjustment for multiple comparisons. In conclusion, second‐ and third‐trimester exposure to vaginal progesterone does not seem to have harmful effects on the childhood health of twins. Lack of adverse maternal events In our previous IPD meta‐analysis47, in which all included studies used vaginal progesterone 90–200 mg/day, the rates of maternal adverse effects, such as vaginal discharge, vaginal pruritus and discontinuation of treatment because of adverse effects, were similar between the vaginal progesterone and placebo groups. In 2013, the three‐armed trial by Serra et al.67 comparing placebo with two different daily doses of vaginal progesterone (200 and 400 mg) reported a dose‐dependent, non‐significant trend towards a higher rate of intrahepatic cholestasis of pregnancy (0% in the placebo group, 1% in the group receiving 200 mg and 5% in the group receiving 400 mg). Nonetheless, the larger study by El‐Refaie et al.69 reported that there was no significant difference in the rate of intrahepatic cholestasis of pregnancy between the group using 400 mg of daily vaginal progesterone (1%) and the no treatment group (0%). Moreover, this study found that the rates of vaginal pruritus, vaginal discharge, headache, skin rash and gastrointestinal symptoms did not differ significantly between the study groups. Thus, it appears that a 400‐mg daily dose of vaginal progesterone is not associated with an increased risk of adverse maternal effects as compared with a 200‐mg daily dose of vaginal progesterone or placebo/no treatment.

che, skin rash and gastrointestinal symptoms did not differ significantly between the study groups. Thus, it appears that a 400‐mg daily dose of vaginal progesterone is not associated with an increased risk of adverse maternal effects as compared with a 200‐mg daily dose of vaginal progesterone or placebo/no treatment. Strengths and limitations The main strengths of our meta‐analysis include: (i) the use of patient‐level data, which offer several advantages over study‐level analysis, including the ability to use more appropriate statistical methods not always feasible using study‐level analysis, define outcome measures consistently across studies, investigate subgroups in which treatment may be either more or less effective, address questions that have not been satisfactorily resolved by individual trials, minimize publication and reporting biases and adjust for prognostic variables that may have confounded the original treatment comparisons; (ii) the baseline balance in prognostic factors between the two study groups, which reduces the possibility of causing bias in the intervention effect estimates; (iii) the absence of substantial heterogeneity in most of the meta‐analyses performed; indeed, all meta‐analyses on the effect of vaginal progesterone on preterm birth had no observed heterogeneity (I 2 = 0%), whereas the majority of meta‐analyses regarding adverse perinatal outcomes had low heterogeneity or no heterogeneity; and (iv) the sensitivity analyses restricted to trials at low risk of bias that were consistent with (and thus supportive of) the overall findings.

reterm birth had no observed heterogeneity (I 2 = 0%), whereas the majority of meta‐analyses regarding adverse perinatal outcomes had low heterogeneity or no heterogeneity; and (iv) the sensitivity analyses restricted to trials at low risk of bias that were consistent with (and thus supportive of) the overall findings. Some potential limitations must also be considered. First, only two trials were specifically designed to assess the efficacy of vaginal progesterone in women with a twin gestation and a sonographic short cervix. Second, 74% of the total sample size of the IPD meta‐analysis was provided by one study69, which included women with a CL between 20 and 25 mm and was not placebo‐controlled. However, it should be highlighted that assessment and measurement of most outcomes included in our review are considered objective in nature, and therefore not likely to be influenced by lack of blinding49. It is noteworthy that estimates of pooled RRs obtained after excluding this study were not significantly different from those obtained in the overall analyses. Moreover, the significant 39% reduction in the risk of composite neonatal morbidity and mortality associated with vaginal progesterone administration was obtained without including data from the study by El‐Refaie et al.69 in the meta‐analysis. Third, the larger study69 did not collect information about several neonatal morbidities, such as necrotizing enterocolitis, intraventricular hemorrhage, proven neonatal sepsis and retinopathy of prematurity. Finally, some subgroup analyses included a small number of patients, which limits the statistical power to estimate the effects within these subgroups.

t information about several neonatal morbidities, such as necrotizing enterocolitis, intraventricular hemorrhage, proven neonatal sepsis and retinopathy of prematurity. Finally, some subgroup analyses included a small number of patients, which limits the statistical power to estimate the effects within these subgroups. Implications for practice and research This updated IPD meta‐analysis indicates that vaginal progesterone reduces the risk of preterm birth and neonatal morbidity and mortality in patients with a twin gestation and a sonographic short cervix, without any deleterious effects on childhood neurodevelopment. Although the results of our meta‐analysis appear promising, further research is required before conclusive advice can be provided with regard to the benefits of using vaginal progesterone in women with a twin gestation and a short cervix. Evidence from this updated IPD meta‐analysis and three ongoing RCTs comparing vaginal progesterone with placebo (NCT02697331 and NCT02518594) or no treatment (NCT02329535) in ∼750 women with a twin gestation and a sonographic short cervix will help to determine whether vaginal progesterone can be recommended to these patients with the aim of preventing preterm birth and improving perinatal outcomes. Supporting information Table S1 Summary of findings of the quality of evidence for each outcome measure Click here for additional data file.

Implications for practice and research This updated IPD meta‐analysis indicates that vaginal progesterone reduces the risk of preterm birth and neonatal morbidity and mortality in patients with a twin gestation and a sonographic short cervix, without any deleterious effects on childhood neurodevelopment. Although the results of our meta‐analysis appear promising, further research is required before conclusive advice can be provided with regard to the benefits of using vaginal progesterone in women with a twin gestation and a short cervix. Evidence from this updated IPD meta‐analysis and three ongoing RCTs comparing vaginal progesterone with placebo (NCT02697331 and NCT02518594) or no treatment (NCT02329535) in ∼750 women with a twin gestation and a sonographic short cervix will help to determine whether vaginal progesterone can be recommended to these patients with the aim of preventing preterm birth and improving perinatal outcomes. Supporting information Table S1 Summary of findings of the quality of evidence for each outcome measure Click here for additional data file. ACKNOWLEDGMENT This research was supported, in part, by the Perinatology Research Branch, Program for Perinatal Research and Obstetrics, Division of Intramural Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services (NICHD/NIH/DHHS); and, in part, with Federal funds from NICHD/NIH/DHHS under Contract No. HHSN275201300006C.

INTRODUCTION Screening for fetal aneuploidy by analysis of cell‐free (cf) DNA in maternal blood was made possible by the advent of massively parallel shotgun sequencing (MPSS) which allows digital counting of cfDNA fragments, either by whole‐genome sequencing1, 2 or targeted approaches3, 4. Detection of fetal trisomy using counting statistics, such as Z‐score or normalized chromosome values (NCV), becomes easier and more robust when the proportion of fetal DNA to total cfDNA in maternal blood is high because of greater separation between normal and aneuploid cases5, 6. The sensitivity of detecting fetal trisomies at low fetal fraction is dependent mostly on the amount of useful counts of the chromosome of interest or sequencing depth6, 7. Inclusion of fetal fraction in analysis algorithms can improve specificity because, in cases of low Z‐scores or NCVs, it helps distinguish between aneuploid cases with low fetal fraction from euploid cases with a higher fetal fraction5. Paired‐end MPSS allows accurate digital counting while also determining the length of each cfDNA fragment by sequencing both its extremities8. As cfDNA fragments of fetal origin are slightly shorter than maternal ones, size differences can be used to determine fetal fraction8. Additionally, in the case of fetal aneuploidy, counting differences detected from all cfDNA fragments would appear more evident if confirmed on shorter fragments only8, 9.

extremities8. As cfDNA fragments of fetal origin are slightly shorter than maternal ones, size differences can be used to determine fetal fraction8. Additionally, in the case of fetal aneuploidy, counting differences detected from all cfDNA fragments would appear more evident if confirmed on shorter fragments only8, 9. neoBona® (Labco Diagnostics, Barcelona, Spain) is the first cfDNA‐based screening test to exploit paired‐end MPSS through a novel bioinformatics approach, which has the advantage of combining conventional counting statistics with the distribution of cfDNA fragment size to provide a double check of chromosome counting data. Additionally, by integrating sequencing depth on each chromosome and fetal fraction it allows calculation of a unique trisomy score (t‐score), thereby quantifying the likelihood of fetal trisomy. The objective of this study was to evaluate the performance of this new method on a large blinded set of archived maternal plasma samples, tested without previous knowledge of their outcomes.

d fetal fraction it allows calculation of a unique trisomy score (t‐score), thereby quantifying the likelihood of fetal trisomy. The objective of this study was to evaluate the performance of this new method on a large blinded set of archived maternal plasma samples, tested without previous knowledge of their outcomes. METHODS Study population Blood samples were collected between April 2006 and February 2015 at King's College Hospital, London, UK, from women with a singleton pregnancy undergoing screening for trisomies 21, 18 and 13 by assessment of a combination of fetal nuchal translucency (NT) thickness and maternal serum free beta human chorionic gonadotropin (β‐hCG) and pregnancy‐associated plasma protein‐A (PAPP‐A) at 11–13 weeks' gestation10. Gestational age was determined from measurement of the fetal crown–rump length11. Women with a high risk from the combined test had chorionic villus sampling (CVS) for fetal karyotyping. Karyotype results obtained from genetic laboratories and details on pregnancy outcome obtained from the maternity computerized records or the general medical practitioners of the women were added into the database as soon as they became available. All patients gave written informed consent to provide samples for research, which was approved by the National Health Service Research Ethics Committee.

tcome obtained from the maternity computerized records or the general medical practitioners of the women were added into the database as soon as they became available. All patients gave written informed consent to provide samples for research, which was approved by the National Health Service Research Ethics Committee. Blood samples were collected into EDTA BD vacutainerTM tubes (Becton Dickinson UK limited, Oxford, UK) and centrifuged at 2000 g for 10 min within 15 min of collection (Plasma 1) followed by another 10 min at 16 000 g to further separate cell debris (Plasma 2). Samples of Plasma 1 and 2 were divided into 0.5‐mL aliquots in separate Eppendorf tubes, labelled with a unique patient identifier and stored at −80 °C for up to 9 years until analysis. A total of 50 cases with trisomy 21, 30 with trisomy 18, 10 with trisomy 13 and 910 normal controls with 1 mL of Plasma 1 or 2 were selected for analysis. Cases with aneuploidy were selected at random and each was matched to 11–12 controls that were sampled on the same or next day as the aneuploid case. Normal controls were uncomplicated pregnancies resulting in live birth after 38 weeks' gestation of phenotypically normal neonates assumed to be euploid. None of the samples was previously thawed and refrozen. Plasma samples (two tubes of 0.5 mL per patient) were coded and sent on dry ice from London to the central laboratory of Labco Diagnostics in Barcelona, Spain, where blinded cfDNA analysis was performed using the neoBona test.

ormal neonates assumed to be euploid. None of the samples was previously thawed and refrozen. Plasma samples (two tubes of 0.5 mL per patient) were coded and sent on dry ice from London to the central laboratory of Labco Diagnostics in Barcelona, Spain, where blinded cfDNA analysis was performed using the neoBona test. Analysis of samples The only information provided to the laboratory for each sample was the patient‐unique identifier, date of collection and whether it was a Plasma 1 or 2 sample. Each sample was assessed for volume, adequacy of labeling and risk of contamination or sample mixing before evaluation of fetal trisomy. Although the volume was < 1 mL (range, 500–950 μL) in 60 of the 1000 samples, they were included in the analysis. Plasma samples from each patient were collected into 96 deep‐well plates. Plates of Plasma 1 samples underwent a second centrifugation step at 16 000 g before DNA extraction. Samples were processed in batches of 96 using VeriSeq NIPT v1.0 chemistry (Illumina Inc, San Diego, CA, USA) on a fully automated workstation (Hamilton Star, Hamilton, Reno, NV, USA) designed to handle plasma isolation, column‐based DNA extraction, set‐up of sequencing library, quantification, normalization and pooling. Sequencing libraries from each batch of 96 samples were collected in two separate pools of 48 double‐indexed samples which underwent paired‐end MPSS for two sets of 36 cycles using NextSeq 500 and 550 sequencers with TG NextSeq 500/550 High Output Kit v1.2 (Illumina inc). Sequencing outputs were analyzed using the VeriSeq NIPT software v1.0 (Illumina inc).

f 96 samples were collected in two separate pools of 48 double‐indexed samples which underwent paired‐end MPSS for two sets of 36 cycles using NextSeq 500 and 550 sequencers with TG NextSeq 500/550 High Output Kit v1.2 (Illumina inc). Sequencing outputs were analyzed using the VeriSeq NIPT software v1.0 (Illumina inc). After de‐multiplexing and filtering, sequence alignment was performed against HG19 for data normalization and interchromosome comparisons7. Regions affected by poor alignment were filtered out and further normalization was applied based on a principal component decomposition as described by Zhao et al.12. Fetal fraction assessment, based on molecular size distributions and differences in coverage between fetal and maternal cfDNA, was complemented with X and Y chromosomes data in cases of male fetuses8, 9, 13. NCVs were calculated for chromosomes 13, 18 and 21, as described previously6, 14. NCV counting statistics are similar in principle to the conventional Z‐score, with a fixed cut‐off of around 3.0 to discriminate between trisomic and unaffected pregnancies, the main difference being that, for NCVs, each chromosome of interest is only normalized against a specific set of chromosomes, optimized for comparable sequencing coverage to minimize variations.

o the conventional Z‐score, with a fixed cut‐off of around 3.0 to discriminate between trisomic and unaffected pregnancies, the main difference being that, for NCVs, each chromosome of interest is only normalized against a specific set of chromosomes, optimized for comparable sequencing coverage to minimize variations. Trisomy likelihood ratios (t‐scores) for each chromosome of interest were calculated for each sample based on the estimated fetal fraction, counting statistics (NCVs) derived from both total and short DNA fragments, and sequencing depth. The likelihood ratio reflects the probability for a sample to be affected, given the observed counting statistics and fetal fraction, versus the probability of a sample to be unaffected, given the same counting data. Thus, using this analysis approach, trisomic samples with low fetal fraction can result in a higher t‐score if they have, for instance, a higher depth of sequencing enabling efficient counting on short DNA fragments which are mostly of fetal origin. Samples were classified as being compatible with the presence or absence of trisomy 21, 18 or 13 using predefined chromosome specific cut‐offs at t‐score values of 1.5 for trisomy 21 and 3.0 for trisomies 18 and 13. Quality‐control analyses (QCs) were applied to monitor sequencing depth, the distribution of cfDNA fragment sizes, sequencing coverage for chromosome denominators and for the estimate of fetal fraction. Results were considered valid only for samples passing all QCs.

Samples were classified as being compatible with the presence or absence of trisomy 21, 18 or 13 using predefined chromosome specific cut‐offs at t‐score values of 1.5 for trisomy 21 and 3.0 for trisomies 18 and 13. Quality‐control analyses (QCs) were applied to monitor sequencing depth, the distribution of cfDNA fragment sizes, sequencing coverage for chromosome denominators and for the estimate of fetal fraction. Results were considered valid only for samples passing all QCs. Results were provided to King's College Hospital in which the classification for each case was compared to pregnancy outcome and detection rates and false‐positive rates were estimated. RESULTS The characteristics of the study population are summarized in Table 1. Compared to euploid pregnancies, in pregnancies with trisomy 21, median maternal age, fetal NT and serum free β‐hCG were higher and serum PAPP‐A was lower and in pregnancies with trisomy 18 or 13 median maternal age and fetal NT were higher and serum free β‐hCG and PAPP‐A were lower. Table 1 Characteristics of 1000 pregnant women undergoing prenatal screening for fetal trisomies, according to outcome

RESULTS The characteristics of the study population are summarized in Table 1. Compared to euploid pregnancies, in pregnancies with trisomy 21, median maternal age, fetal NT and serum free β‐hCG were higher and serum PAPP‐A was lower and in pregnancies with trisomy 18 or 13 median maternal age and fetal NT were higher and serum free β‐hCG and PAPP‐A were lower. Table 1 Characteristics of 1000 pregnant women undergoing prenatal screening for fetal trisomies, according to outcome Characteristic Euploid (n = 910) Trisomy 21 (n = 50) Trisomy 18 (n = 30) Trisomy 13 (n = 10) Maternal age (years) 31.9 (27.3–34.9) 37.9 (35.3–41.3) 35.4 (28.8–40.5) 33.5 (30.0–34.9) Maternal weight (kg) 65.0 (59.0–75.0) 68.0 (60.7–75.0) 66.8 (60.4–75.5) 64.5 (60.6–64.9) Maternal height (cm) 165 (160–169) 166 (161–172) 166 (160–171) 167 (164–170) Racial origin Caucasian 563 (61.9) 41 (82.0) 17 (56.7) 8 (80.0) Afro‐Caribbean 244 (26.8) 6 (12.0) 6 (20.0) 1 (10.0) South Asian 32 (3.5) 1 (2.0) 3 (10.0) 0 (0) East Asian 26 (2.9) 2 (4.0) 2 (6.7) 0 (0) Mixed 45 (4.9) 0 (0) 2 (6.7) 1 (10.0) Cigarette smoker 55 (6.0) 3 (6.0) 1 (3.3) 1 (10.0) Method of conception Spontaneous 880 (96.7) 46 (92.0) 26 (86.7) 10 (100) Ovulation drugs 9 (1.0) 3 (6.0) 2 (6.7) 0 (0)

6 (12.0) 6 (20.0) 1 (10.0) South Asian 32 (3.5) 1 (2.0) 3 (10.0) 0 (0) East Asian 26 (2.9) 2 (4.0) 2 (6.7) 0 (0) Mixed 45 (4.9) 0 (0) 2 (6.7) 1 (10.0) Cigarette smoker 55 (6.0) 3 (6.0) 1 (3.3) 1 (10.0) Method of conception Spontaneous 880 (96.7) 46 (92.0) 26 (86.7) 10 (100) Ovulation drugs 9 (1.0) 3 (6.0) 2 (6.7) 0 (0) In‐vitro fertilization 21 (2.3) 1 (2.0) 2 (6.7) 0 (0) Fetal crown–rump length (mm) 61.8 (57.0–67.6) 66.1 (60.0–73.0) 56.1 (51.9–61.6) 59.0 (51.1–63.1) GA at screening (weeks) 12.6 (12.2–13.0) 12.9 (12.5–13.4) 12.2 (11.8–12.6) 12.4 (11.8–12.7) Fetal NT thickness (mm) 1.7 (1.5–1.9) 4.4 (3.4–6.2) 6.5 (3.6–7.9) 5.2 (2.3–6.3) PAPP‐A MoM 1.126 (0.766–1.563) 0.695 (0.441–0.869) 0.227 (0.135–0.327) 0.371 (0.282–0.570) Free β‐hCG MoM 0.995 (0.678–1.582) 2.259 (1.574–3.109) 0.293 (0.171–0.362) 0.314 (0.204–0.747) Data are given as median (interquartile range) or n (%). β‐hCG, beta human chorionic gonadotropin; GA, gestational age; MoM, multiples of the median; NT, nuchal translucency; PAPP‐A, pregnancy‐associated plasma protein‐A. The cfDNA test provided results for 988 (98.8%) cases. In total, 12 (1.2%) samples, nine from euploid and three from trisomy 21 pregnancies, failed to provide a result and were excluded from further analysis. The reasons for QC failure were size distribution of cfDNA fragments beyond the expected range (n = 6), low sequencing depth for the observed fetal fraction (n = 4), unusually high DNA concentration (n = 1) and insufficient sequencing coverage for determination of fetal fraction (n = 1).

excluded from further analysis. The reasons for QC failure were size distribution of cfDNA fragments beyond the expected range (n = 6), low sequencing depth for the observed fetal fraction (n = 4), unusually high DNA concentration (n = 1) and insufficient sequencing coverage for determination of fetal fraction (n = 1). The cfDNA test classified correctly all 47 pregnancies with fetal trisomy 21, all 10 with trisomy 13, 29 (96.7%) of 30 with trisomy 18 and all 901 unaffected pregnancies (Table 2). In one case of trisomy 18, t‐score and NCV values for chromosome 18 were compatible with normal chromosome copy number; in this case the fetal fraction was 11%. One case with trisomy 21 and one unaffected pregnancy had the same NCV of 3.5, but had different t‐scores for trisomy 21 which were 10 and −14, respectively. Therefore, using the predefined cut‐offs of t‐score values of 1.5 for trisomy 21 and 3.0 for both trisomies 18 and 13 resulted in detection rates of 100% for trisomies 21 and 13 and 96.7% for trisomy 18, with false‐positive rate of 0% for all trisomies. Table 2 Results of cell‐free DNA analysis by neoBona® test for fetal trisomy screening in 988 women with test result, according to outcome Result Euploid (n = 901) Trisomy 21 (n = 47) Trisomy 18 (n = 30) Trisomy 13 (n = 10) NCV for chromosome 21 −0.01 (−3.43 to 3.55) 11.50 (3.59 to 25.67) 0.32 (−1.88 to 3.48) −0.048 (−1.36 to 1.25)

Table 2 Results of cell‐free DNA analysis by neoBona® test for fetal trisomy screening in 988 women with test result, according to outcome Result Euploid (n = 901) Trisomy 21 (n = 47) Trisomy 18 (n = 30) Trisomy 13 (n = 10) NCV for chromosome 21 −0.01 (−3.43 to 3.55) 11.50 (3.59 to 25.67) 0.32 (−1.88 to 3.48) −0.048 (−1.36 to 1.25) t‐score for trisomy 21 −23.2 (−1074.2 to 0.6) 101.0 (7.2 to 392.1) −12.0 (−178.4 to −1.1) −20.5 (−121.8 to −5.2) NCV for chromosome 18 0.01 (−3.25 to 6.17) −0.24 (−2.42 to 2.71) 12.24 (−1.22* to 36.91) 0.72 (−1.49 to 2.56) t‐score for trisomy 18 −31.3 (−1960.6 to 1.0) −38.6 (−322.2 to −3.6) 94.5 (−17.9* to 765.2) −21.9 (−247.2 to −7.2) NCV for chromosome 13 0.001 (−4.55 to 4.44) −0.09 (−1.91 to 2.58) 0.42 (−2.30 to 2.16) 14.76 (6.31 to 28.50) t‐score for trisomy 13 −37.6 (−2591.6 to 0.1) −48.6 (−449.6 to −1.3) −15.2 (−442.6 to 2.0) 209.3 (23.9 to 479.5) Fetal fraction (%) 10.2 (0.3 to 33.8) 10.7 (3.8 to 19.8) 9.6 (0.8 to 23.0) 7.9 (4.0 to 15.3) Trisomy 21 t‐score > 1.5 0 (0) 47 (100) 0 (0) 0 (0) Trisomy 18 t‐score > 3.0 0 (0) 0 (0) 29 (96.7) 0 (0) Trisomy 13 t‐score > 3.0 0 (0) 0 (0) 0 (0) 10 (100) Data are given as median (range) or n (%). * Value from the only discrepant result; case of trisomy 18 with 11% fetal fraction resulted in trisomy likelihood score (t‐score) and normalized chromosome value (NCV) compatible with normal chromosome 18 copy number.

t‐score for trisomy 13 −37.6 (−2591.6 to 0.1) −48.6 (−449.6 to −1.3) −15.2 (−442.6 to 2.0) 209.3 (23.9 to 479.5) Fetal fraction (%) 10.2 (0.3 to 33.8) 10.7 (3.8 to 19.8) 9.6 (0.8 to 23.0) 7.9 (4.0 to 15.3) Trisomy 21 t‐score > 1.5 0 (0) 47 (100) 0 (0) 0 (0) Trisomy 18 t‐score > 3.0 0 (0) 0 (0) 29 (96.7) 0 (0) Trisomy 13 t‐score > 3.0 0 (0) 0 (0) 0 (0) 10 (100) Data are given as median (range) or n (%). * Value from the only discrepant result; case of trisomy 18 with 11% fetal fraction resulted in trisomy likelihood score (t‐score) and normalized chromosome value (NCV) compatible with normal chromosome 18 copy number. The mean fetal fraction was 10.6% for euploid pregnancies, 11.1% for trisomy 21, 9.4% for trisomy 18 and 8.9% for trisomy 13. One case of trisomy 21, three of trisomy 18 and 58 unaffected pregnancies were identified correctly despite showing fetal fractions below 4%, including one case of trisomy 18 and nine euploid cases with fetal fraction < 1% (Table 2 and Figure 1).

ncies, 11.1% for trisomy 21, 9.4% for trisomy 18 and 8.9% for trisomy 13. One case of trisomy 21, three of trisomy 18 and 58 unaffected pregnancies were identified correctly despite showing fetal fractions below 4%, including one case of trisomy 18 and nine euploid cases with fetal fraction < 1% (Table 2 and Figure 1). Figure 1 Normalized chromosome value (NCV) (a) and trisomy likelihood score (t‐score), for values between −100 and 100 (b), for chromosome 18, in 29 pregnancies with trisomy 18 () and 901 unaffected pregnancies (), plotted against fetal fraction. Plot of NCV in (a) shows limitation of this method because, in two cases of trisomy 18 with low fetal fraction (< 4%) (arrows in (a) and (b)), NCV was similar to those of two unaffected cases (circled) with high fetal fraction; these unaffected cases would have been classified wrongly as positive for trisomy 18 as they are above the cut‐off of 3.0 (). Plot of t‐scores in (b) shows cases of trisomy 18 with score well above cut‐off of 3.0 for trisomy 18 (), including one case with fetal fraction < 1%.

d cases (circled) with high fetal fraction; these unaffected cases would have been classified wrongly as positive for trisomy 18 as they are above the cut‐off of 3.0 (). Plot of t‐scores in (b) shows cases of trisomy 18 with score well above cut‐off of 3.0 for trisomy 18 (), including one case with fetal fraction < 1%. UOG-17386-FIG-0001-cDISCUSSION The findings of this study demonstrate the feasibility of a new approach for cfDNA testing of maternal blood in screening for fetal trisomies 21, 18 and 13. Paired‐end MPSS of cfDNA coupled with a novel analysis algorithm provided simultaneous assessment of fetal fraction, distribution of size of DNA fragments and chromosome counting. Trisomy likelihood ratios for each chromosome of interest could then be calculated for each sample based on the estimated fetal fraction, chromosome‐specific counting statistics on total and short fragments and sequencing depth. We used this novel approach to examine stored plasma samples and, at preselected chromosome‐specific cut‐offs of t‐score values of 1.5 for trisomy 21 and 3.0 for trisomies 18 and 13, the test classified correctly all cases of trisomy 21, trisomy 13 and unaffected pregnancies and 29 of 30 cases of trisomy 18. Such high performance of screening is compatible with the best results of previous studies utilizing cfDNA testing to screen for trisomies 21, 18 and 1315.

y 21 and 3.0 for trisomies 18 and 13, the test classified correctly all cases of trisomy 21, trisomy 13 and unaffected pregnancies and 29 of 30 cases of trisomy 18. Such high performance of screening is compatible with the best results of previous studies utilizing cfDNA testing to screen for trisomies 21, 18 and 1315. In the single case of trisomy 18 that was misclassified, the fetal fraction was 11% and is therefore highly unlikely that this error was related to technical issues affecting test sensitivity. Unfortunately, no more sample was available to repeat the test and exclude errors due to laboratory mishandling. In addition, trisomy rescue, generating a normal cell line in the cytotrophoblast, could not be ruled out as the underlying cause of this discrepancy as prenatal diagnosis was only performed on long‐term CVS culture by quantitative fluorescent polymerase chain reaction and karyotype, but not on direct preparation.

ing. In addition, trisomy rescue, generating a normal cell line in the cytotrophoblast, could not be ruled out as the underlying cause of this discrepancy as prenatal diagnosis was only performed on long‐term CVS culture by quantitative fluorescent polymerase chain reaction and karyotype, but not on direct preparation. The basis for cfDNA testing using counting methods is that, in trisomic pregnancies, the number of molecules derived from the extra fetal chromosome, as a proportion of all sequenced molecules in maternal plasma, is higher than in euploid pregnancies. The ability to detect the small increase in the amount of a given chromosome in maternal plasma in a trisomic compared to a disomic pregnancy is related directly to the fetal fraction and the depth of sequencing3, 16, 17, 18, 19. Trisomy cases with low fetal fraction used to be more difficult to discriminate from normal samples by counting statistics only, as they can produce NCVs with similar values to those occasionally observed in normal samples with higher fetal fraction6, thus reducing test specificity. Also the sensitivity could be affected if, for the sequencing depth used, the proportion of fetal cfDNA is too low to allow discrimination of trisomies by counting statistics only7, 18. For these reasons, when the fetal fraction is below 4%, which occurs in 0.5–6.1% of pregnancies, the cfDNA test is usually presented as a failure and no result is reported15.

f, for the sequencing depth used, the proportion of fetal cfDNA is too low to allow discrimination of trisomies by counting statistics only7, 18. For these reasons, when the fetal fraction is below 4%, which occurs in 0.5–6.1% of pregnancies, the cfDNA test is usually presented as a failure and no result is reported15. Some of the problems due to low fetal fraction have now been overcome by the application of the multicomponent t‐score, as the resolution in discriminating between trisomic and unaffected pregnancies is no longer dependent only on fetal fraction but also on the new possibility of performing additional counting statistics on short DNA fragments, which are mostly of fetal origin. Consequently, trisomic pregnancies with low fetal fraction could result in higher t‐score values than in pregnancies with higher fetal fraction and lower total sequencing depth or less efficient counting statistics on short fragments. This approach proved to be highly efficient at low fetal cfDNA amounts, as all four aneuploid cases with fetal fraction between 0.8% and 3.5% could be detected. The effectiveness of the new multicomponent t‐score to improve overall specificity was evident in one case of trisomy 21 and one unaffected sample that were classified correctly despite generating the same NCV, and thus would be undistinguishable by conventional MPSS analysis algorithms.

.8% and 3.5% could be detected. The effectiveness of the new multicomponent t‐score to improve overall specificity was evident in one case of trisomy 21 and one unaffected sample that were classified correctly despite generating the same NCV, and thus would be undistinguishable by conventional MPSS analysis algorithms. Despite testing archived plasma samples, and with suboptimal volumes in 6% of cases, failure to provide a result was only observed in 1.2% of samples. The most common reason for test failure was an abnormal distribution of size of cfDNA fragments, which affected size‐based counting and the measurement of fetal fraction. This artifact was likely to be caused by cfDNA shearing, resulting from sample degradation. It is therefore expected to occur less frequently in routine clinical samples collected in dedicated tubes, designed to prevent cell lysis and stabilize cfDNA. Four more samples failed the QC analysis which combines sequencing metrics and estimated fetal fraction, thus determining if the analysis output has statistical confidence in scoring a sample. Repeating the test on a second aliquot of the same plasma would probably have yielded a valid result. In clinical routine, the test is usually performed within a few days from sampling and with enough volume to be repeated, therefore this technical failure is expected to decrease.

stical confidence in scoring a sample. Repeating the test on a second aliquot of the same plasma would probably have yielded a valid result. In clinical routine, the test is usually performed within a few days from sampling and with enough volume to be repeated, therefore this technical failure is expected to decrease. The novel approach presented in this study has the potential of extending the advantages of cfDNA‐based aneuploidy screening to a wider proportion of pregnancies. Complementing conventional counting statistics with size‐based chromosome counting and fetal fraction ensured that accurate prediction of trisomic status was provided in 62 of our cases with fetal fraction < 4%. Consequently, it is no longer necessary to exclude samples from analysis solely because the fetal fraction is < 4% if enough sequencing depth is reached for the corresponding amount of cfDNA and size‐based counting is performed at the same time. ACKNOWLEDGMENTS This study was supported by a grant from The Fetal Medicine Foundation (UK Charity No: 1037116). The neoBona test was performed by Labco Diagnostics, Barcelona, Spain with support from Illumina Inc. (San Diego, CA, USA).

INTRODUCTION Diagnosis of fetal aneuploidy relies on invasive testing by chorionic villus sampling (CVS) or amniocentesis in pregnancies that are identified by screening to be at high risk for such aneuploidies1. In many developed countries the most widely accepted method of screening at present is one that is based on a combination of maternal age, fetal nuchal translucency (NT) thickness, maternal serum free beta‐human chorionic gonadotropin (β‐hCG) and pregnancy‐associated plasma protein‐A (PAPP‐A), that could identify approximately 90% of fetuses with trisomies 21, 18 or 13 at a false‐positive rate (FPR) of 5%2. A new method of screening for fetal trisomy relies on the examination of cell‐free DNA (cfDNA) in maternal plasma3. Several studies in the last 4 years have reported the clinical validation and/or implementation of cfDNA testing and a recent meta‐analysis of such studies weighted pooled detection rates (DR) and FPR in singleton pregnancies as 99.2% and 0.09%, respectively, for trisomy 21, 96.3% and 0.13% for trisomy 18, and 91.0% and 0.13% for trisomy 134. The IONA® test and IONA software (Premaitha Health plc, Manchester, UK) is a cfDNA test developed recently which uses the Ion Proton™ sequencing platform (Thermo Fisher Scientific, Waltham, MA, USA) and an algorithm that determines the relative number of chromosomal copies, enabling the detection of fetal trisomies5. The objective of this study was to assess the potential performance of the IONA test and IONA software in screening for trisomies 21, 18 and 13 at 11–13 weeks' gestation.

Scientific, Waltham, MA, USA) and an algorithm that determines the relative number of chromosomal copies, enabling the detection of fetal trisomies5. The objective of this study was to assess the potential performance of the IONA test and IONA software in screening for trisomies 21, 18 and 13 at 11–13 weeks' gestation. METHODS This was a nested case–control study of stored maternal plasma from 242 singleton pregnancies at 11–13 weeks' gestation, including 201 with euploid fetuses, 35 with trisomy 21, four with trisomy 18 and two with trisomy 13. In all cases fetal karyotyping was performed following CVS in our tertiary referral center because screening by the combined test had demonstrated an increased risk for trisomies 21, 18 or 13. Gestational age was determined from the measurement of fetal crown–rump length (CRL)6. Maternal venous blood (10 mL) was collected before CVS in ethylenediaminetetraacetic acid vacutainer tubes (Becton Dickinson UK Limited, Oxfordshire, UK) and was processed within 15 min of collection. Samples were centrifuged at 2000 g for 10 min to separate the plasma from packed cells and buffy coat (plasma 1) and again at 16 000 g for 10 min to further separate cell debris (plasma 2). Plasma 1 and 2 (2 mL each) were divided into 0.5‐mL aliquots in separate Eppendorf tubes that were labeled with a unique patient identifier and stored at –80°C until subsequent analysis. Written informed consent was obtained from the women who agreed to participate in the study, which was approved by the King's College Hospital Ethics Committee.

were divided into 0.5‐mL aliquots in separate Eppendorf tubes that were labeled with a unique patient identifier and stored at –80°C until subsequent analysis. Written informed consent was obtained from the women who agreed to participate in the study, which was approved by the King's College Hospital Ethics Committee. We searched our database and selected 35 consecutive cases of trisomy 21, four cases of trisomy 18 and two cases of trisomy 13 that had 2 mL of stored plasma 2 available. Two hundred and one euploid control subjects were selected; none of their samples was previously thawed and refrozen. Maternal blood was collected between April 2007 and June 2012. Laboratory analysis Plasma samples (four tubes of 0.5 mL per patient) from selected cases were sent from London to the laboratory of Premaitha Health plc in Manchester, UK. The following information was provided to Premaitha for each case: patient‐unique identifier, maternal age, weight and height, racial origin, method of conception, smoking habit, gestational age in weeks, fetal CRL, and date of blood collection. Before evaluation for fetal trisomy, Premaitha assessed each sample for volume, adequacy of labeling, and risk of contamination or sample mixing and informed us that all samples met their acceptance criteria. The 242 samples were then analyzed using the IONA® test5. Results were provided for the risk of trisomies 21, 18 and 13 in each case and the correlation between the assay results and the fetal karyotype was determined.

isk of contamination or sample mixing and informed us that all samples met their acceptance criteria. The 242 samples were then analyzed using the IONA® test5. Results were provided for the risk of trisomies 21, 18 and 13 in each case and the correlation between the assay results and the fetal karyotype was determined. Statistical analysis Descriptive statistics are presented as median (interquartile range (IQR)) for continuous variables and n (%) for categorical variables. Probability scores for each trisomy are presented in scatterplots. DR and FPR are reported based on a predefined cut‐off for probability. An age‐adjusted probability of trisomy calculated to be ≥ 1 in 150 was considered a positive result, as used in the UK National Health Service (NHS)7. Point estimates with 95% CI have been provided. RESULTS All 242 samples were processed with the IONA test, however, one of these did not meet the validity criteria applied by the IONA software owing to to a low fetal fraction and was excluded from subsequent analysis. In this case, sampling was performed at 13 weeks' gestation, the maternal weight was 86 kg and the serum PAPP‐A multiples of the median (MoM) value was 1.316 and there were no obvious reasons for why the fetal fraction was low. Results were generated by the IONA software for the remaining 241 samples.

sequent analysis. In this case, sampling was performed at 13 weeks' gestation, the maternal weight was 86 kg and the serum PAPP‐A multiples of the median (MoM) value was 1.316 and there were no obvious reasons for why the fetal fraction was low. Results were generated by the IONA software for the remaining 241 samples. The characteristics of the euploid and aneuploid pregnancies are summarized in Table 1. In all 35 cases of trisomy 21, the probability score for trisomy 21 was > 95% and the scores for trisomies 18 and 13 were ≤ 0.0001% (Figure 1). In all four cases of trisomy 18, the probability score for trisomy 18 was > 77% and the scores for trisomies 21 and 13 were ≤ 0.0001%. In the two cases of trisomy 13, the probability score for trisomy 13 was > 59% and the scores for trisomies 21 and 18 were ≤ 0.0001%. In the 200 euploid pregnancies with a test result, the probability score was < 0.08% for trisomy 21, < 0.001% for trisomy 18 and < 0.002% for trisomy 13. Therefore, the DR for trisomy 21 was 100% (95% CI, 90.1–100.0%; 35/35 cases), the DR for trisomy 18 was 100% (95% CI, 51.0–100.0%; 4/4 cases) and the DR for trisomy 13 was 100% (95% CI, 34.2–100.0%; 2/2 cases), with a FPR of 0% (95% CI, 0.0–1.9%; 200/200 cases). Table 1 Characteristics of study population of women with a singleton pregnancy who underwent cell‐free DNA analysis using the IONA® test, grouped according to karyotype obtained following chorionic villus sampling

The characteristics of the euploid and aneuploid pregnancies are summarized in Table 1. In all 35 cases of trisomy 21, the probability score for trisomy 21 was > 95% and the scores for trisomies 18 and 13 were ≤ 0.0001% (Figure 1). In all four cases of trisomy 18, the probability score for trisomy 18 was > 77% and the scores for trisomies 21 and 13 were ≤ 0.0001%. In the two cases of trisomy 13, the probability score for trisomy 13 was > 59% and the scores for trisomies 21 and 18 were ≤ 0.0001%. In the 200 euploid pregnancies with a test result, the probability score was < 0.08% for trisomy 21, < 0.001% for trisomy 18 and < 0.002% for trisomy 13. Therefore, the DR for trisomy 21 was 100% (95% CI, 90.1–100.0%; 35/35 cases), the DR for trisomy 18 was 100% (95% CI, 51.0–100.0%; 4/4 cases) and the DR for trisomy 13 was 100% (95% CI, 34.2–100.0%; 2/2 cases), with a FPR of 0% (95% CI, 0.0–1.9%; 200/200 cases). Table 1 Characteristics of study population of women with a singleton pregnancy who underwent cell‐free DNA analysis using the IONA® test, grouped according to karyotype obtained following chorionic villus sampling Euploid Trisomy 21 Trisomy 18 Trisomy 13 Characteristic (n = 200) (n = 35) (n = 4) (n = 2) Maternal age (years) 32.9 (29.2–36.9) 38.0 (35.5–40.5) 37.7 (30.0–40.1) 32.1 (31.3–32.1) Maternal weight (kg) 67.0 (60.0–76.0) 63.6 (57.2–74.5) 68.5 (60.9–72.0) 53.6 (50.8–53.6) Maternal height (cm) 165 (160–170) 165 (160–167) 172 (166–173) 163 (162–163) GA at testing (weeks) 12.7 (12.2–13.1) 13.0 (12.4–13.6) 11.9 (11.4–13.3) 12.8 (12.0–12.8) Racial origin Caucasian 146 (73.0) 31 (88.6) 2 (50.0) 2 (100.0) Afro‐Caribbean 38 (19.0) 1 (2.9) 1 (25.0) 0 (0) South Asian 5 (2.5) 3 (8.6) 0 (0) 0 (0) East Asian 3 (1.5) 0 (0) 0 (0) 0 (0) Mixed 8 (4.0) 0 (0) 1 (25.0) 0 (0) Cigarette smoker 10 (5.0) 3 (8.6) 1 (25.0) 0 (0) Mode of conception Spontaneous 189 (94.5) 33 (94.3) 2 (50.0) 2 (100.0) Ovulation drugs 2 (1.0) 2 (5.7) 0 (0) 0 (0)

o‐Caribbean 38 (19.0) 1 (2.9) 1 (25.0) 0 (0) South Asian 5 (2.5) 3 (8.6) 0 (0) 0 (0) East Asian 3 (1.5) 0 (0) 0 (0) 0 (0) Mixed 8 (4.0) 0 (0) 1 (25.0) 0 (0) Cigarette smoker 10 (5.0) 3 (8.6) 1 (25.0) 0 (0) Mode of conception Spontaneous 189 (94.5) 33 (94.3) 2 (50.0) 2 (100.0) Ovulation drugs 2 (1.0) 2 (5.7) 0 (0) 0 (0) In‐vitro fertilization 9 (4.5) 0 (0) 2 (50.0) 0 (0) Fetal NT (mm) 1.8 (1.5–2.2) 4.2 (3.1–5.3) 2.5 (1.6–6.8) 5.9 (2.8–5.9) Free β‐hCG MoM 1.232 (0.821–1.950) 2.551 (1.696–3.615) 0.352 (0.224–0.901) 1.136 (0.515–1.136) PAPP‐A MoM 1.144 (0.736–1.612) 0.708 (0.500–1.157) 0.232 (0.145–0.288) 0.621 (0.146–0.621) Data are given as median (interquartile range) or n (%). β‐hCG, beta‐human chorionic gonadotropin; GA, gestational age; MoM, multiples of the median; NT, nuchal translucency; PAPP‐A, pregnancy‐associated plasma protein‐A. Figure 1 Probability scores for trisomy 21 (a), trisomy 18 (b) and trisomy 13 (c) in singleton pregnancies with trisomic () or euploid () fetuses that underwent cell‐free DNA analysis using the IONA® test. UOG-15749-FIG-0001-bDISCUSSION Main findings of the study This nested case–control study has demonstrated that, in pregnancies considered by the combined test to be at high risk for trisomies 21, 18 or 13, cfDNA testing of maternal plasma at 11–13 weeks' gestation using the IONA test identified correctly all trisomic pregnancies, at a FPR of 0%.

ION Main findings of the study This nested case–control study has demonstrated that, in pregnancies considered by the combined test to be at high risk for trisomies 21, 18 or 13, cfDNA testing of maternal plasma at 11–13 weeks' gestation using the IONA test identified correctly all trisomic pregnancies, at a FPR of 0%. Comparison with findings from previous studies The performance of the IONA test in the detection of trisomy 21 is comparable to the reported performance of cfDNA analysis in a recent meta‐analysis4. Although the number of cases of trisomies 18 and 13 is too small for definitive conclusions to be drawn, the performance of the test may be higher than in other studies8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30. In this study, compared with most previous publications on cfDNA testing, we used samples that were obtained at 11–13 weeks exclusively. This is important because, in the last decade, there has been a major shift from second‐ to first‐trimester screening and diagnosis of aneuploidies.

2, 23, 24, 25, 26, 27, 28, 29, 30. In this study, compared with most previous publications on cfDNA testing, we used samples that were obtained at 11–13 weeks exclusively. This is important because, in the last decade, there has been a major shift from second‐ to first‐trimester screening and diagnosis of aneuploidies. Strengths and limitations of the study The study has several strengths, including a large number of pregnancies affected by trisomy 21, laboratory staff blinded to the fetal karyotype or outcome at the time of cfDNA testing, and complete outcome data. The study examined the application of a novel technique and software. The main limitations include its retrospective design, inclusion of only high‐risk pregnancies and a small number of pregnancies affected by trisomies 18 or 13. However, the reported performance of cfDNA testing is similar in studies in both high‐ and low‐risk populations11, 22, 23, 25, 29, 31, 32, 33, 34, we therefore believe that the results of the study can be applied to a general population but in an ideal situation such results should be validated in a prospective study. Conclusion In this study, the IONA test differentiated all cases of trisomy 21, 18 and 13 from euploid pregnancies. ACKNOWLEDGMENTS This study was supported by a grant from The Fetal Medicine Foundation (UK Charity No: 1037116). The IONA® test was provided by Premaitha Health plc, Manchester, UK.